Lecture Notes, Advanced Artificial Intelligence (SCOM7341) Sina Institute, University of Birzeit 2 nd Semester, 2012 Advanced Artificial Intelligence (SCOM7341) Chapter 2 Intelligent Agents Dr. Mustafa Jarrar Sina Institute, University of Birzeit mjarrar@birzeit.edu www.jarrar.info Jarrar 2012 1
Outline Agents and environments Rationality PEAS (Performance measure, Environment, Actuators, Sensors) Environment types Agent types Jarrar 2012 2
Agents An agent is anything that can be viewed as perceiving its environment through sensors and acting upon that environment through actuators Human agent: eyes, ears, and other organs for sensors; hands. legs, mouth, and other body parts for actuators Robotic agent: cameras and infrared range finders for sensors. various motors for actuators Jarrar 2012 3
Agents and environments The agent function maps from percept histories to actions: [f: P* A] The agent program runs on the physical architecture to produce f agent = architecture + program Jarrar 2012 4
Vacuum-cleaner world Percepts: location and contents, e.g., [A,Dirty] Actions: Left, Right, Suck, DoNothing Jarrar 2012 5
A vacuum-cleaner Agent Tabulation of an agent function of the vacuum-cleaner Jarrar 2012 6
Rational Agents An agent should strive to "do the right thing", based on what it can perceive and the actions it can perform. The right action is the one that will cause the agent to be most successful. Performance measure: An objective criterion for success of an agent's behavior. E.g., performance measure of a vacuum-cleaner agent could be amount of dirt cleaned up, amount of time taken, amount of electricity consumed, amount of noise generated, etc. Jarrar 2012 7
Rational Agents Rational Agent: For each possible percept sequence, a rational agent should select an action that is expected to maximize its performance measure, given the evidence provided by the percept sequence and whatever built-in knowledge the agent has. Jarrar 2012 8
Rational Agents Rationality is distinct from omniscience (all-knowing with infinite knowledge) Agents can perform actions in order to modify future percepts so as to obtain useful information (information gathering, exploration) An agent is autonomous if its behavior is determined by its own experience (with ability to learn and adapt) Jarrar 2012 9
PEAS When designing a rational/intelligent agent, we keep in mind PEAS PEAS: Performance measure, Environment, Actuators, Sensors Consider, e.g., the task of designing an automated taxi driver: Performance measure Environment Actuators Sensors Jarrar 2012 10
PEAS Agent: automated taxi driver Performance measure: Safe, fast, legal, comfortable trip, maximize profits Environment: Roads, other traffic, people and objects in/around the street Actuators: Steering wheel, accelerator, brake, signal, horn Sensors: Cameras, sonar, speedometer, GPS, odometer, engine sensors, keyboard Jarrar 2012 11
PEAS Agent: Medical diagnosis system Performance measure: Healthy patient, minimize costs, lawsuits Environment: Patient, hospital, staff Actuators: Screen display (questions, tests, diagnoses, treatments, referrals) Sensors: Keyboard (entry of symptoms, findings, patient's answers) Jarrar 2012 12
PEAS Agent: Part-picking robot Performance measure: Percentage of parts in correct bins Environment: Conveyor belt with parts, bins Actuators: Jointed arm and hand Sensors: Camera, joint angle sensors Jarrar 2012 13
PEAS Agent: Interactive English tutor Performance measure: Maximize student's score on test Environment: Set of students Actuators: Screen display (exercises, suggestions, corrections) Sensors: Keyboard Jarrar 2012 14
Environment Types Fully observable (vs. partially observable): An agent's sensors can measure all relevant aspects of the environment at each point in time. Deterministic (vs. stochastic): The next state of the environment is completely determined by the current state and the action executed by the agent. Vacuum is deterministic? Taxi driver is stochastic? (If the environment is deterministic except for the actions of other agents, then the environment is strategic). Episodic (vs. sequential): The agent's experience is divided into atomic "episodes" (each episode consists of the agent perceiving and then performing a single action), and the choice of action in each episode depends only on the episode itself. Chess and taxi driver are sequential because the current action affect the next action. Jarrar 2012 15
Environment Types Static (vs. dynamic): The environment is unchanged while an agent is deliberating. (The environment is semidynamic if the environment itself does not change with the passage of time but the agent's performance score does). Taxi driver is dynamic, chess is static, chess with clock I semidynamic. Discrete (vs. continuous): A limited number of distinct, clearly defined percepts and actions. Chess has a finite number of distinct states, thus it is discrete; however the Taxi-driving is not. Single agent (vs. multiagent): An agent operating by itself in an environment. Crossword is Single, while Chess is a two-player environment. Jarrar 2012 16
Environment Types Chess with Chess without Taxi driving a clock a clock Fully observable Yes Yes No Deterministic Strategic Strategic No Episodic No No No Static Semi Yes No Discrete Yes Yes No Single agent No No No The environment type largely determines the agent design The real world is (of course) partially observable, stochastic, sequential, dynamic, continuous, multi-agent Jarrar 2012 17
Agent Functions & Programs An agent is completely specified by the agent function mapping percept sequences to actions One agent function (or a small equivalence class) is rational Aim: find a way to implement the rational agent function concisely Jarrar 2012 18
Agent Types Four basic types in order of increasing generality: Simple reflex agents Model-based reflex agents Goal-based agents Utility-based agents Jarrar 2012 19
Simple Reflex Agents The agent selects an action(s) based on the current precept, ignoring the rest of the precept history. Jarrar 2012 20
Model-based Reflex Agents The agent decides its action(s) based on a predefined set of condition-action rules. A telephone operator/answering machine. Jarrar 2012 21
Goal-based Agents The agent decides its action(s) based on a known goal. For example, a GPS system finding a path to certain destination. Jarrar 2012 22
Utility-based Agents The agent decides its action(s) based on utilities/preferences. a GPS system finding a shortest/fastest/safer path to certain destination. Jarrar 2012 23
Learning Agents The agent adapts its action(s) based on feedback (not only sensors). Jarrar 2012 24
Discussion Panel Summarize the new knowledge you have really learned about agents? Do you really agree that it is possible to realize such agents, or it is only another name for programs? What is the difference between an agent and a program? When an agent is not a program? When the program is not an agent. Jarrar 2012 25